Abstract Our accelerating ability to measure biological systems at the molecular, cellular, organ and organism level promises a new generation of powerful and improved drug therapies. However, drug failures continue to occur at high rates due to lack of efficacy or unexpected toxicity?even when the drug binds its intended target with very high affinity. We do not sufficiently understand the biological systems in which we are intervening, suggesting that we are making fundamental assumptions that are wrong. Building on results from our previous grant period, this renewal proposal proposes new assumptions: (a) when drugs work it is because they interact not only with their target but with many other off-targets that produce synergistic effects, (b) the actions of drugs can be best understand as the interaction between protein networks that are dysfunctional in disease and drug response networks that are modulated by the complete set of relevant targets, and (c) that evidence of direct physical interaction is superior to complicated and integrative signals (such as gene expression) in creating and analyzing drug response networks that can usefully be linked to disease networks. Thus, we propose a plan to (1) develop and apply methods to predict drug interactions on a proteome scale, and uses these to improve methods for creating interaction networks relevant to drug response and disease biology, (2) devise methods to associate drug response with disease biology, using the features of the associated protein networks, and (3) collaborative apply these tools with collaborations from academia (U. Pennsylvania for NSAID response & the Structural Genomics Consortium for target selection and triage), industry (Genentech for cancer, Pfizer for autoimmune disease), and government (the U.S. FDA for seeking biomarkers to predict efficacy and toxicity. With success, we will have created a framework for drug discovery, repurposing, combination use and toxicity prediction that may contribute to a higher rate of success in delivering new therapies to benefit public health.